Senior Engineer 2: AI Agentic Solutions (hybrid - Seattle, Wa)

Nordstrom Nordstrom · Retail · Seattle, WA

Senior Engineer role focused on designing and building AI agentic solutions end-to-end for Nordstrom. This involves agent engineering, context engineering, evaluations, guardrails, and memory/state management. The role requires hands-on experience with LLMs, agentic frameworks, RAG, and production deployment, with a product-minded approach and strong software engineering skills.

What you'd actually do

  1. Partner with business and technology stakeholders to define the “art of the possible” with agents — translating ambiguous problems into agentic solutions with clear success criteria and measurable outcomes.
  2. Design and build core agentic solutions end-to-end across orchestration, tool-use pipelines, and integration with enterprise systems.
  3. Own end-to-end solution design for agentic solutions spanning multiple engineers’ work, with full upstream/downstream integration consideration.
  4. Apply context engineering to determine what an agent sees, when, and why — balancing token economics, latency, and decision quality across RAG patterns, structured retrieval, and dynamic prompt assembly.
  5. Develop and own evaluations and guardrails that demonstrate solutions are safe, reliable, and accurate — offline benchmarks, online production telemetry, and failure-mode analysis.

Skills

Required

  • 6+ years of professional software engineering experience, with a strong track record of designing and delivering complex, scalable distributed systems.
  • Hands-on experience working with LLMs, foundation model APIs (OpenAI, Anthropic, Google, etc.), prompt engineering, retrieval-augmented generation (RAG) architectures, and embedding-based search in production environments.
  • Experience designing, building, and operating AI agents or agentic workflows in production, including tool-use, orchestration, and integration with downstream systems.
  • Strong understanding of how to assemble, prune, and structure context for agents to maximize decision quality within token, latency, and cost constraints.
  • Experience designing evaluation frameworks and safety guardrails for LLM-based systems, including offline benchmarks, online telemetry, and responsible deployment practices.
  • Familiarity with short-term and long-term memory patterns for agents, vector stores, conversation state, and durable workflow state.
  • Hands-on experience with agentic frameworks such as Claude Agent SDK, LangGraph, AutoGen, CrewAI, Semantic Kernel, or OpenAI Assistants API.
  • Familiarity with multi-agent orchestration patterns: task decomposition, tool-use pipelines, and human-in-the-loop workflows.
  • A product-minded approach to engineering: strong instincts for user impact, comfortable pushing back on requirements when the right solution isn’t the one initially asked for, and able to translate business intent into agentic capabilities.
  • Proficiency in Python; strong grasp of multiple tech stacks and cloud-native development on AWS and/or GCP.
  • Experience working with cross-functional teams including product, business, infrastructure, and security stakeholders.
  • Strong verbal and written communication skills; ability to articulate complex technical decisions to both technical and non-technical audiences.
  • Agile development experience (Scrum, Kanban, Lean, or similar) with a continuous improvement and quality mindset.
  • Bachelor’s or Master’s degree in Computer Science, Engineering, or equivalent pract

Nice to have

  • AI Agentic Solutions team
  • context engineering
  • memory and state management for agentic solutions
  • orchestration
  • tool-use pipelines
  • integration with enterprise systems
  • RAG patterns
  • structured retrieval
  • dynamic prompt assembly
  • offline benchmarks
  • online production telemetry
  • failure-mode analysis
  • short-term context
  • long-term memory
  • durable conversation state
  • LLM APIs
  • embedding models
  • vector stores
  • agentic frameworks
  • Claude Agent SDK
  • LangGraph
  • AutoGen
  • CrewAI
  • Semantic Kernel
  • OpenAI Assistants API
  • multi-agent orchestration patterns
  • task decomposition
  • human-in-the-loop workflows
  • AWS
  • GCP

What the JD emphasized

  • AI Fluency — Required
  • Hands-on experience working with LLMs, foundation model APIs (OpenAI, Anthropic, Google, etc.), prompt engineering, retrieval-augmented generation (RAG) architectures, and embedding-based search in production environments.
  • Experience designing, building, and operating AI agents or agentic workflows in production, including tool-use, orchestration, and integration with downstream systems.
  • Experience designing evaluation frameworks and safety guardrails for LLM-based systems, including offline benchmarks, online telemetry, and responsible deployment practices.
  • Hands-on experience with agentic frameworks such as Claude Agent SDK, LangGraph, AutoGen, CrewAI, Semantic Kernel, or OpenAI Assistants API.

Other signals

  • AI Agentic Solutions team
  • designing and building those solutions end-to-end
  • agent engineering
  • evaluations and guardrails
  • memory and state management for agentic solutions
  • product-minded engineer
  • AI Fluency
  • AI agents or agentic workflows in production
  • agentic frameworks